Abstract: A recommendation system uses artificial intelligence to identify, based on negative sentiment cues from users, item attributes, such as keywords, that users may find offensive or undesirable. The negative sentiment cues may be explicit (e.g., a user selects an option not to view a particular recommendation again), implicit (e.g., a user does not interact with recommendations relating to an attribute), or both. The system may use a computer model generated based on these identified attributes to filter or modify recommendations to a user or group of users. For instance, if a particular keyword is identified as highly offensive to a group of users, items associated with the keyword may be filtered from item recommendations presented to the group of users. If an attribute is identified as moderately offensive to a user, items associated with the attribute may be down-weighted in item recommendations presented to the user.

Abstract: A mechanism is disclosed for evaluating the quality of a machine translation system. Propositions are identified within a source language string and within a translation of the source language string generated by a machine translation system. The propositions can be identified using an automated mechanism or using human translators through the use of a human intelligence task site. Once the propositions have been identified, the propositions identified in the source language string can be compared to the propositions identified in the translated target language string. The results of the comparison can be utilized to compute a quality score for the translation. A final quality score can be generated for the machine translation system by repeating this process for multiple source language strings. The final quality score can then be utilized to improve the quality of models utilized by the machine translation system.

Abstract: A computing resource service provider may provide computing instances organized in to logical groups, such as auto-scaling groups. Computing instances assigned to an auto-scaling group may be detached from auto-scaling group. Detached instances may no longer be managed by the auto-scaling group and may not contribute to the capacity of the auto-scaling group for auto-scaling purposes.

Abstract: Computing resource service providers allow customers to execute computer systems on hardware provided by the computing resource service provider. The hardware may be distributed between various geographic locations connected by a network. A load balancer may be provided to distribute traffic between the computer systems. Furthermore, computing resource service provider may cause computing resources to be allocated or deallocated to the load balancer based at least in part on various attributes of the computer systems the load balancer is responsible for distributing traffic to. The various attributes may include a capacity of the computer systems.

Abstract: Technologies are disclosed for voice-controlled monitoring of computing resources in a service provider network. Verbal requests for information about the status of computing resources in a service provider network are recorded and transmitted to a voice service. The spoken utterances are translated into text and analyzed to determine the intent of the speaker and to extract relevant words by a voice service. The voice service provides a request to a serverless compute service that identifies the intent of the speaker and includes the words extracted from the verbal utterance. The serverless compute service, in turn, executes a voice command processing function to identify the resources, obtain the status information, and construct a textual reply to the request that includes the status information. The textual reply can then be provided to the voice service for conversion into spoken audio and playback to the requesting user.

Abstract: Source code of a graph analysis program expressed in a platform-independent language which supports linear algebra primitives is obtained. An executable version of the program is generated, which includes an invocation of a function of a parallel programming library optimized for a particular hardware platform. A result of executing the program is stored.

Abstract: A method and system for providing language services to a client system in a web-based code development environment. In one embodiment, the client system includes a local data store or cache to store code completion information received from a web service. The code completion information corresponds to an identified scope of the code set. Upon detection of the input of a newly entered character in the same scope of the code set, the code completion information is retrieved from the cache of the client system. The retrieved code completion information may be used to complete the entry of a code element beginning with the newly entered character.

Abstract: Systems and methods are described for testing computing resources. In one embodiment, a search space of computing settings is analyzed in accordance with weighted data that maps computing performance parameters with the computing settings. A subset of the computing settings is selected to generate a test population to optimize at least one computing performance parameter. One or more computing devices in a computing environment are configured in accordance with the test population, and the test conditions are iteratively updated based on test results in accordance with the test population and a fitness function.

Abstract: Systems, methods, and computer-readable media are disclosed for systems and methods for dynamic mode switching and management of communications between devices. Example methods include receiving a first event from a first application on a first device, determining a first application identifier of the first application, and determining that the first device is in a communal mode. Example methods may include determining a communal mode profile for the first device, where the communal mode profile is associated with a first user account identifier that is associated with the first device, determining a set of user account identifiers associated with an accessory device identifier of the accessory device, and associating the set of user account identifiers and the accessory device identifier with the communal mode profile. Example methods may include receiving an indication of a second event, and disassociating the set of user account identifiers from the communal mode profile.

Abstract: A system and method for operating an automated aerial vehicle are provided wherein influences of ground effects (e.g., which may increase the effective thrusts of propellers by interfering with the respective airflows) are utilized for sensing the ground or other surfaces. In various implementations, operating parameters of the automated aerial vehicle are monitored to determine when ground effects are influencing the parameters associated with each of the propellers, which correspondingly indicate proximities to a surface (e.g., the ground). Utilizing such techniques, proximities of different portions of an automated aerial vehicle to the ground or other surfaces may be determined (e.g., for detecting issues with an uneven landing area, a sloped ground, etc.).

Abstract: Techniques are described for decoding portions of image frames for virtual reality (VR) media content. A field of view of a viewer within a VR environment can be determined and used to decode a portion of an image frame that provides image content for that portion. The other portions of the image frame that are not within the image frame can remain non-decoded when not in the visible field of view.

Abstract: Generally described, systems and methods are provided for detecting the impact of network failures. The system collects performance information from a plurality of nodes and links in a network, aggregates the collected performance information across paths in the network, processes the aggregated performance information for detecting failures on the paths, adjusts the set of performance information by removing the performance information for any nodes considered to be associated with performance information that is statistically different from performance information from other nodes at a given location or extrapolates the collected information to other paths, and determines the impact to customers of the network failures detected using the adjusted set of performance information.

Abstract: Systems and processes are described for establishing and using a secure channel. A shared secret may be used for authentication of session initiation messages as well as for generation of a private/public key pair for the session. A number of ways of agreeing on the shared secret are described and include pre-sharing the keys, reliance on a key management system, or via a token mechanism that uses a third entity to manage authentication, for example. In some instances, the third party may also perform endpoint selection by providing a particular endpoint along with the token. The particular cipher suite applied in a particular implementation may be configurable. The process is applicable to either implicit key confirmation (e.g., handshake negotiation) or explicit key confirmation (e.g., full negotiation).

Abstract: A physical host agent receives configuration information from a virtual computer system service specifying network traffic information to be extracted from network traffic for one or more virtual machines. The agent extracts the specified network traffic information from the network traffic for the one or more virtual machines and aggregates the network traffic information into one or more data segments for storage in a repository. A publishing sub-system of the service obtains the one or more data segments and compiles the one or more data segments into data logs for delivery to an analytics service to make the network traffic information available to customers.

Abstract: Described is a system and method for propagating, synchronizing and expiring user specific information among multiple output devices. In some implementations, retention information is provided along with the user specific information and the output device and/or output device controller will determine when to expire the user specific information based on the retention information. In other implementations, the output device controller and/or output device may periodically request updates from other output devices, output device controllers and/or the inventory management system to determine whether to expire user specific information.

Abstract: A recommendation system uses artificial intelligence to identify, based on negative sentiment cues from users, item attributes, such as keywords, that users may find offensive or undesirable. The negative sentiment cues may be explicit (e.g., a user selects an option not to view a particular recommendation again), implicit (e.g., a user does not interact with recommendations relating to an attribute), or both. The system may use a computer model generated based on these identified attributes to filter or modify recommendations to a user or group of users. For instance, if a particular keyword is identified as highly offensive to a group of users, items associated with the keyword may be filtered from item recommendations presented to the group of users. If an attribute is identified as moderately offensive to a user, items associated with the attribute may be down-weighted in item recommendations presented to the user.

Abstract: A method for processing packet data in a service provider environment includes, by a network-enabled data processing device within a server computer of the service provider environment, receiving packet data comprising header information and payload information. The header information is separated from the payload information. The separated header information is forwarded to a processor of the server computer for processing, without forwarding at least a portion of the payload information. At least one address of a storage location is received from the processor. The at least one address is associated with a logical-to-physical address mapping based on the header information. The payload information is stored in the storage location based on the at least one address.

Abstract: Requests are pre-generated to include a cryptographic key to be used in fulfilling the requests. The requests may be encoded in uniform resource locators and may include authentication information to enable a service provider to whom the requests are submitted to determine whether the requests are authorized. The requests may be passed to various entities who can then submit the requests to the service provider. The service provider, upon receipt of a request, can verify the authentication information and fulfill the request using a cryptographic key encoded in the request.

Abstract: Systems and methods are provided for learning and using a linear machine learning model, where the training data includes a categorical variable. After encoding the categorical variable, a computing system can learn a linear machine learning model. From the linear machine learning model, the computing system can construct a numeric variable. The numeric variable can be used in computing a prediction result and in learning a new linear machine learning model.

Abstract: This disclosure describes a configuration of an aerial vehicle, such as an unmanned aerial vehicle, in which one or more of the propellers are positioned within a duct that includes an active airflow channel within the interior of the duct. The active airflow channel actively moves within the duct so that it remains aligned with the tips of the blades of the propeller within the duct. As the propeller and the active airflow channel rotate, at least some of the airflow structures (e.g., vortices) shed from the blades of the propeller are collected by the active airflow channel and channeled away from the propeller so that a following blade of the propeller does not pass through the collected airflow structures.